A previous question/answer here mentioned Mozart's ability (when he was a teenager) to sit and listen to a performance of Allegri's Miserere, remember most of the details, and then write it down later that day.

Is this really possible? Even for a genius like Mozart, all the different voices, the length of the piece, and the fact that he only heard it once—not to mention his age—makes it seem a bit unrealistic that he could remember, let's say 75%, and then write that down from memory with less than, let's say 25%, of error.

My question:

  1. Is this really possible? Or it is likely that someone with him (his father) helped also, or they were writing it down as they were listening? Even if he had help or was able to transcribe while listening, it still seems unrealistic.

  2. Assuming this is possible, how exactly does one achieve this ability (even remotely close)? I know ear training is a huge thing, and that with time and practice you will be able to dictate, however how can one listen to an extended piece of music (more than 10 minutes) and be able to transcribe all voices/instruments in a manner that is at least, let's say, 75% accurate?

Wiki page here

EDIT: Thanks everyone for the great discussion.

  • 13
    As an aside, for one of our gigs I have had to get a session guitarist in to take my place as I'll be away. Now I know my music is much simpler, but this guy listened to 10 songs straight through and wrote them down after one listening. Almost note perfect! And he is not a savant, just a professional with many years experience as a session musician. – Doktor Mayhem May 20 '15 at 7:02
  • 8
    Human memory in the age of smartphones is not what it used to be. In the days of Beowulf and The Odyssey, heroic feats of recall were considered somewhat normal. It seems reasonable that it could apply to music as well, for a trained musician. – 200_success May 20 '15 at 17:42
  • 2
    Read Luria's "The mind of a mnemonist". docs.google.com/file/d/… – osa May 21 '15 at 3:20
  • Not quite the same thing, but at the 2015 Proms, an entire orchestra will be playing Beethoven's 6th from memory. They did the same for Mozart's 40th last year. The human mind is a remarkable thing. – AakashM May 21 '15 at 8:37
  • @200_success Such epics were recited (or rather, composed from a stock set of phrases on the fly) by trained professionals. It's a stretch to assume that the ability to reproduce, word for word, pieces of that length after one hearing was ever considered "common". – chepner May 21 '15 at 18:59

Neuroscience still can't explain some of the amazing things human brains can do. A person alive in our time has interesting and similar ability to mentally manage music in a way that suggests that some people might be wired for this sort of thing.

The study detailed at http://www.radiolab.org/story/148670-4-track-mind/ showed that Bob Milne has the ability to mentally play up to four symphonies in his head simultaneously. Early in his career he didn't use or need sheet music - he could listen to someone play a piece, then play it back in complete.

So to be specific:

Is this really possible?

Yes. It's a rare ability, but there are those that can memorize a long song with one listening session, and repeat it themselves or copy it to sheet music.

how exactly does one achieve this ability?

Lots of practice would probably do it. Start with short one hand pieces, then lengthen them, then add more voices. There are many techniques for memorization. Some will find it easier, some harder, depending on skills and talents they've already developed in their life, but I don't think there's any specific reason why one couldn't train their brain for this type of work.

  • Lots of practice would probably do it. - Really? I don't see any suggestion that this ability could be trained. You can most certainly improve your memory etc., but this kind of eidetic memory seems to be a one in a million genetic quirk. – Davor May 21 '15 at 12:47
  • 3
    @Davor I'm not claiming that one could definitely become like Milne through practice. I'm claiming that the central talent posed, "sit and listen to a performance of Allegri's Miserere, remember most of the details, and then write it down later that day." is something that, with sufficient practice, could be learned by some, if not most musicians. Milne simply shows that people can do far, far more than "simple" recall, so doing the recall, even at a young age, is possible, though we can't verify easily whether Mozart himself did it. Can such extreme memory be learned? I don't know. – Adam Davis May 21 '15 at 12:58
  • 2
    However, since ear training is a common practice in many educational institutions, and memory training is common, then it's reasonable to conclude that not only is it possible to learn "simple" recall and recording, but many musicians go through significant effort to improve their skills in this area. I expect that if there were no results, and everyone left such training disappointed, then it wouldn't continue to be pursued. en.wikipedia.org/wiki/Ear_training – Adam Davis May 21 '15 at 13:01

I wasn't there, but I would not find it completely out of the question. In every field of expertise, experts are capable of chunking information in ways that amateurs are not.

An expert listener will not just hear a few hundred notes performed by several voices, they will hear harmonies, their relationships to each other, and rhythmical patterns. More importantly, they will be able to represent the vast majority of those patterns very compactly, because they follow typical progressions, allowing them to focus on the parts that are atypical (and it's not as if renaissance music was wildly experimental).

Somebody like Mozart would be highly familiar with the typical patterns of this music, so it's not impossible that he would have been able to memorize the entire piece. If he had, instead, heard a piece of 20th century vocal music like this, he might have had much greater difficulties.

  • 1
    I always heard the legend was that he did this at the age of 4, making it a bit more incredible, but what fun are legends if they don't strain credulity? – Darren Ringer May 19 '15 at 23:00
  • 1
    That makes a lot of sense, and it is what I was thinking. Still though, let's assume that Mozart was able to recognize and remember the harmonic progressions 100%. This still would not allow him to recognize and remember all of the inner voices... Maybe chunking this information would result in him guesstimating the inner voices? – lobi May 19 '15 at 23:08
  • 4
    I once met a teenager who had been taught the piano largely without sheetmusic. When we tried to play some 4 handed stuff from sheetmusic, it was a desaster, until she asked me to play just her part once, and suddenly she had it down perfectly. We later explored this a bit and I played some fugues with 3 voices, which she reproduced almost flawlessly from hearing them once. We tried fugues with 4 voices, where she got maybe 75% correct. I know Miserere is not a fugue, but I think it's easier than a fugue. So yes, I think it's possible even for a teenager. – Sumyrda May 20 '15 at 6:07
  • 3
    @lobi I always just assumed that if he even got half the inner voices right most listeners would not know the difference, and then spend the rest of their lives spreading rumors about how perfect the transcription was. He probably got younger with every iteration, too. – Darren Ringer May 20 '15 at 15:28

It's worth noting that the Miserere is extremely repetitive. For example, this version is roughly 15 minutes long, but you get all the melodic and harmonic content in the first 2:45 except for the final cadence; everything after that is more verses set to the same music. Mozart still would have had to remember the varying text overlay as well as any ornamentation that changed from verse to verse, but transcribing the Miserere from memory would have been a lot easier than transcribing a random 15-minute piece, even one in an idiom he was familiar with.

Which is not to say that the story is necessarily true, but it does make it more plausible...


All documented resources I can find agree with the story. The Pope, instead of excommunicating Mozart, conferred on him the Order of the Golden Spur, a papal knighthood, for "contributing to the glory of the Church" through his transcription of the Miserere. The record of that award, as well as the minutes of that audience, are part of the archives of the Catholic Church in Rome.

As far as how he was able to do so, it's quite a feat, but it's probably an oversimplification of things to say that he heard a 12-minute piece of choral music, then went home and wrote the whole thing down from memory:

  • Many sources say Mozart was able to sit in at a rehearsal as well as the Good Friday Tenebrae performance, so he got two bites at the apple, and might even have been able to sneak a peek at the sanctioned manuscript. Popular myth says he actually drafted a copy after this rehearsal and smuggled it into the Tenebrae service for corrections, but this account is discredited.
  • The piece is twelve minutes long, but is very repetitive, with each of the three main subgroups of the choir (main choir, cantors and quartet) having the same melody each time they sing. So, Mozart would have gotten several chances to hear each melodic line even if he only heard the full piece once.
  • It was very probable that he had access to a Latin Bible for the text of Psalm 51, if he didn't already know the text from memory. It's a staple of the Catholic Tenebrae service on Good Friday, so Mozart would very likely have heard or even recited it at least every Good Friday for years before hearing Allegri's setting of it.
  • He also possibly had access to an instrument in his guest quarters in Rome that he could use to test what he was transcribing.
  • Lastly, what Allegri originally wrote was a very simple "falsobordone" recitative work, designed to be ornamented by the Chapel Choir soloists, and choristers slowly added these abbellimenti for over 150 years, teaching the variations to new members by rote memory, before Mozart heard the Sistine Chapel Choir perform it in 1770. So, it's impossible to know exactly what Mozart heard and therefore how close he came to an exact reproduction.
  • It doesn't help that Mozart's manuscript was "sanitized" to remove these abbellimenti before Burney was allowed to publish it; the ornamentation added by the Chapel Choir would remain a "trade secret" in the Vatican for another 70 years, before Pietro Alfieri published a detailed account in 1840 of what the Chapel Choir had actually been singing this whole time. This later set of abbellimenti, with possible further embellishment from what Mozart had heard, would be (incorrectly) woven into various transcriptions leading up to Haas' 1931 "top C" variant, which we consider the "authoritative" version of the work for modern performance.

All told, it's not like you or I hearing something like Beethoven's Fifth in a concert hall and then going home and writing out the full orchestral score from memory, perfectly. It's more akin to writing out the guitar tab of your favorite rock song... having seen it played only once or twice... without a recording to play against while you write.. and coming up with something close enough but probably missing a couple flourishes. Still quite an accomplishment, especially for a 17-year-old.


I agree with what microtherion says:

Most professional classical pianists who practice pieces during all their lives end up remembering an outstanding succession of notes and rhythm and this is mostly due to patterns (physical, visual, melodic, etc.) that help organize and make sense of what they recall.

If you were to present a professionally trained pianist with a random succession of notes, they would have a much more challenging task in remembering the piece even if it were only a couple of pages long.

Now as for Mozart and the Misere, just a couple of points I can only quote from memory:

  • I do not believe there are written traces of Mozart's accomplishment with the Misere piece, but only an account from his father.

  • I also read that his editor did embellish a number of details in his life to boost up his image and sells.

I don't think anyone can question Mozart's genius but it does not prevent us from doubting the veracity of some of the tales, and the Misere one is among those that are perhaps most vague and questionable.

A plausible scenario for example is that he listened to the piece, recalled pretty accurately the overall progression and structure of the piece, recomposed/filled in a number of gaps guided by the musical theory concepts in effect at the time, corrected some details upon listening to it a few more times, and ended up with an interpretation close enough to the original that it would sound very similar to the original to most people. An accomplishment few people--especially children--could do.

  • 2
    I share this healthy scepticism. In the version of the tale I was told (verbally), all of the scores were collected after the performance and it was not published. How could an independent person verify Mozart's writing of it was absolutely note perfect? No doubt he could remember every important detail of the piece, but a note-perfect copy wouldn't be necessary. – Andy May 20 '15 at 8:31
  • 1
    As Andy says, unless there was another person who was similarly talented, how would that transcription be authenticated? Even if the original performers heard it, they probably wouldn't know how accurate it was. – Tim May 20 '15 at 11:49
  • @Andy Excellent point. – lobi May 20 '15 at 19:46

A past iteration of the Wikipedia page on Mozart's compositional method had a good summary of the mythologization that took place regarding Mozart in the 19th century. (The fact that that page has been substantially reworked is a testament to the extent to which that process is still contentious today.)

Note, the following is an old wikipedia link


Another instance of Mozart's powerful memory concerns his memorization and transcription of Gregorio Allegri's "Miserere" in the Sistine Chapel as a 14-year-old. Here again, various factors suggest great skill on Mozart's part, but not a superhuman miracle: the work in question is somewhat repetitive, and that Mozart was able to return to hear another performance, correcting his earlier errors. Solomon suggests that Mozart may have seen another copy earlier.[17]

Essentially, it's entirely possible he did memorize it in one or two listenings (though there seems at least some possibility that he 'cheated' as well).

The resources being:

Solomon, Maynard (1995) Mozart: A Life. New York: Harper Perennial.

Irving, John (2006) "Sonatas," in Cliff Eisen and Simon P. Keefe, The Cambridge Mozart Encyclopedia. Cambridge: Cambridge University Press.

  • Welcome to the site Vr Rm! Great first answer, complete with citations! – Josiah May 22 '15 at 0:56

I had a professor tell us about a comp lesson with Igor Stravinsky once. (I know right?!?) He paged once, slowly through the piece my professor brought at his desk then moved to the piano and started playing excerpts pointing out places that could have been different and why they might work better.... leaving the score at his desk. So I'd say totally possible.


Other posters have pretty much captured it, I think, but it might help to think of yourself retelling a joke you just heard. Chances are you don't remember it word-for-word, and if you were trying to recall it that way, you'd fail. You do remember the basic shape and form of the joke (probably not least because it shares the shape and form in common with many other jokes you've told), and you remember the way the various parts have to link together to lead to a punchline. Most of us are pretty capable of re-telling a joke that we might have only heard once.

In music, a trained ear can pick out similar structures. You may have to remember a few bits and pieces along the way, but if you have a broad understanding of how music tends to work, then you can join the dots between those key points. Mozart was listening to music which followed established formulae, and he was very familar with the techniques for iterating those formulae. Had he jumped in a time machine and listened to the Rite of Spring, things might have been different.


To adress question 2: I don't know if it can be learned. It was natural for me.

(I'm the guy mentioned in the accepted answer, so this personal account is perhaps relevant – thanks for your kind words, everyone. A friend sent me your link.)

  • Wow Bob Milne responded to my question! Amazing! (assuming you are in fact Bob Milne...) – lobi Jul 1 '15 at 18:38
  • 1
    Yes, I am. Check my website Www.bobmilne.com Or else download a new thing they've made for me called an app. It's free on the app stores. I'm not very good with sites like this. – Bob Milne Jul 1 '15 at 23:06
  • 1
    Welcome anyway, and sorry that we're a bit pedantic here – only, it does help to keep the site clean of garbage posts. I tried to convert your comment into an actual answer to the question, hopefully it is Ok now. – leftaroundabout Jul 4 '15 at 15:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.